Measuring speaker response with a computer (1 Viewer)

[Greg->G]

Hi,

I'm interested in using a computer to measure the response of my speakers, instead of doing it manually with a ratshack spl meter. It seems like it would be much easier to tweak something, then just hit the Go button and have the computer generate all the tones and take the measurements, and maybe draw a graph.

I'm actually surprised that I haven't seen more people do something like this on this forum. Can anybody tell me why this is not more common? Is it the cost of the software neccessary?

My real question is: what hardware do I need to make this work reasonably well? Specifically, what kind of microphone. I received a microphone with my receiver for its auto-calibration (pioneer elite 45tx). Would that do? I also saw that there is an output plug on the ratshack spl meter. Could I use that if the mic from my receiver isn't good enough?

Just for reference, I plan to use whatever setup I come up with for setting up a subwoofer as soon as I get one (moving into a house soon, can't wait!).

Thanks,
greg

 

[Jason Dalton]

Well, cost is a big factor. I've been (casually) looking into the same thing, I can't say I know a whole lot about it or that I have any recommendations but I know it will likely cost you a few hundred dollars for software and equipment. Thats a lot more expensive than a RS meter and a freeware tone generator.

 

[Greg->G]

Well, the thing for me is that I have access to software that I think would work for no cost. That's the big reason I'm thinking about pursuing this. Also, I'm a software developer, so if I can't get usable software from work, I might consider writing my own. I don't think it would be especially complicated (but I could be way off on that).

 

[Johnny_M]

also consider that every microphone is not made equal. any resistance in the mic is going to give you different readings from 1 mic to another. i would imagine the cost of the propper microphone is going to cost you close to what youd spend on the spl meter.

i was in your shoes about 2 months ago, i just recently moved from a small apt to a large house and was so excited about finally setting up my HT properly and being able to listen at the levels i wanted to. i was hesitant to spend the 50 bucks on a spl meter but i finally talked myself into it, and im EXTREMELY happy that i did. that is probably the best money ive spent on my ht so far. just knowing thats its being done properly without any doubt in your head makes it so much easier to enjoy a movie, no more thinkin, is the right speaker louder than the left? etc...

also consider this, in my case i calibrated using test tones from a dvd player, my receiver does not have independant speaker adjustments for each input, so when i tested the calibration again from the receiver internal tones, they did not match, so you might find if you calibrate with your computer, your dvd input may be off. unless your receiver is better than mine.

Johnny

 

[ChrisWiggles]

When you measure your "speakers" with a rat shack meter and test tones, you are not measuring your speakers at all, but you are also very heavily measuring the impact of your room acoustics. There are methods to measure your speakers that approximate an anechoic chamber using computer software which would much more accurately reflect only the speakers, and not the *huge* influence the room has.

You can start exploring one such method here if you want: http://www.owlnet.rice.edu/~elec301/...1/TDS/tds.html

 

[David Judah]

This one is popular because it is easy to use and very thorough--you can download a trial version if you want to try it out. Alot of members use it here. Try a search under "ETF" and you should get alot of threads about it and other measurement software.

Alot of programs do have compensation for different microphones. It's good not to scrimp on the mic.

DJ

 

[Greg->G]

ChrisWiggles,

Actually I did mean that I wanted to measure my speakers combined with room response, and get an idea of how flat (or un-flat, but hopefully not) my response curve is at my listening position(s).

 

[Jimi C]

i just downloaded and installed that ETF program to check it out and now im done with it and i can not get sound out of my pc speakers.

 

[Kevin C Brown]

ETF5. I used this to great satisfaction to eq my sub. Very excellent software.

 

[VinhT]

I originally purchased TrueRTA for the purpose of measuring the response of my subwoofer in my room. Used in conjuction with the Radio Shack SPL meter, it provided quick, consistent results and I was satisfied for quite a while. Then, the time came for serious equalization work in my vehicle, which mandated a better microphone (within the means of a college student). So, I purchased the calibrated mic/preamp combo from ETF Acoustic. The system works fine, and while I use it primarily for car audio applications, being able to see what my HT speakers are doing in-room has proven to be extremely useful for proper setup and calibration.

 

[Greg->G]

Thanks for your feedback everyone! I think I am going to get the rat shack spl meter to use as a microphone with whatever I end up using, and I am going to try etf. It looks like it will do exactly what I want. I downloaded the demo and looked at it, and it looks good. Does anyone know if the demo has any limitations or time limit or anything? I didn't see any details about that on their website. Thanks again for your help everyone!

 

[Mark Seaton]

One of the best separate mic's for inexpensive measurement is the Mity Mic from AudioeXpress or the solution sold by the providers of ETF. For those working with outboard mic-preamps, the best cost effective solution I have found to be reliable is one of the 1/2" condenser microphone from Superlux. It can be had for under $100 and typicaly measures within +/-0.5-1dB of super expensive measurement microphones.

There are many different software measurement packages. TrueRTA is quite good for subwoofer integration and EQ, and ETF and WinMLS are great options for a wide range of testing, and there are many others.

Most spend more on cabling for their system than what some basic measurement equipment costs. Also consider that this is one purchase which will give you endless ability to tweak, experiment, and truly improve your system's performance. This is certainly a good and valuable investment for those with upgrade-itis.

 

[Kevin C Brown]

Maek and Vinh- You ever compare a calibrated microphone with the Radio Shack meter? Just wondering if getting something separate would be worth it. < $100 ain't bad though.

 

[MingL]

ETF is a good software for transient response measurement. Spectraplus or truerta are good for steady state measurements.

I usually use both when I want meticulous results, otherwise using spectraplus or truerta would typically give the type of measurements associated with an "automatic" ratshack measure/charting combo.

 

[Greg->G]

Pardon my ignorance, but what is the difference between transient response and steady state measurements? I haven't heard either of those terms before.

 

[ChrisWiggles]

Transient response is usually like the first bit of a sound, or how "fast" things can react to create sounds. For instance, hitting a drum has very significant and strong transients at impact, its a very abrupt change. A flute, however, playing a sustained tone would be more a steady state tone, and the transients at the beginning of the tone, while they would exist, would be far less jarring than that of percussive instruments/sounds.

 

[Greg->G]

So which of these two is better for calibrating a sub within a room? It seems like the steady state would be the correct choice. Am I correct in assuming that?

 

[MingL]

Each method has its own good and bad points. I tend to do steady state more often because it closely approximates what we would do with a spl meter and a disc of test tones.

I suggest jumping into steady state measurements first. Familiarise yourself with this measurements and when the time is right and curiousity gets the better of you, you will venture onto transient measurements.

Otherwise, both methods will get the same approximate freq response of your system in the room its in.

 

[Kevin C Brown]

Steady state has a tendency of exciting and sustaining room nodes. Transients are so quick that they really don't. I would think both are important for a sub. Easier to do steady state though.